Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[dirkson]

Wow! You guys post a lot when I sleep : )

Although I would also add that SlocCount has never been a good estimate of anything* and so advertising that it's an excellent estimate of the work you've done so far was the main reason I disregarded your kickstarter.

As your posting here I'll probably take another look as a dev that follows up peoples interest is a good sign imo, but really you want concrete showing what it can do examples to show how much work you've put in.

I did change the wording there, removing the "excellent" and the money-mention. I think that's improved the section a little. While, given, line count is a sucky metric, it's actually possible to count and talk about - And, more importantly, it's one of the few metrics that's somewhat accessible to the layman. The phrase "We have no non-negligable memory leaks" would leave a programmer happy, but would terribly confuse your average person.

Actually, I forgot this, but in the kickstarter there's a 44m youtube video of someone building a "frankkinship", which should be Frankenship. He shows the heat engine working, the voxel-based damage, organic regeneration and a few other things. There's already a fairly large amount of work.

I do suck as describing my game. Time and time again, other people managed to do it better than I can. I was so incredibly thankful for that video, which shows off a ton of stuff.

Regarding the ego, I think it's the fact you assume that you are very interesting to other people. The tip-off for this, to me, was the "seriously" in the end of your full bio, when the description strikes me (at least) as perfectly normal. [...] I hope you don't take this the wrong way.

I like what you are saying. In your opinion, why do you think that the game is already so slow? Is it the fact that it's not block base but voxel based?

You must know a lot more tweed-wearing cavers than I do! Nothing about my description is particularly /amazing/, but a lot of the hobbies I pick up are not particularly common. Oh, and you'll have to work a lot harder to offend me!

The game is actually a lot /faster/ than it used to be, and getting faster. But yes, the reason this is a problem is the fact that it's voxel-based. To render the equivalent of a single minecraft block, we have to deal with 4096 cubes. This scales rapidly - The average scene needs to be able to work out which of 4billion+ cubes need to be displayed roughly 15 times per second.

Which is a huge thing, since there's absolutely infinite amounts of data to send (like the temperatures and whatnot).

Is the architecture already in a client/server model, as MC changed it to?

It's not yet client/server, but it's being designed with that switch in mind.

There's a TON of data that exists... But the trick is that most of that data either doesn't need to exist on others' computers, only needs to exist in a tiny sphere (Where the player can see), or can be re-calculated on the fly from more simple data. (I.e. A voxel hole would take hundreds of bytes if I sent it directly - Or < 10 bytes if I merely send the event that CAUSED the voxel hole)

A combination of these methods should make for reasonable data use. My goal is to make it work with the internet connection I have - 3mb/s down, <1mb/s up.

Shields

I think I'll make shields magnetic fields - This seems like the most plausible real-world phenomenon that corresponds to sci-fi shields.

Why would space-ship ramming be less effective than water-ship ramming?

Speed and mass.

A water-ship was to be, on average, less dense than the water it's floating in. Our largest aircraft carriers can displace over 100,000 tons of water. This means, if I'm remembering my physics correctly, (I may not be!) that they'd be 100,000 tons themselves. Spaceships don't need to me any more or less dense than anything, and adding on mass from asteroids is easy. Thus they attain larger tonnages rather quickly. By happy coincidence, I happen to have a partially-build ship roughly the size of an aircraft carrier. It masses 2,500,000 tons, and it's just an empty shell.

Our fastest large water-ships run at around 30 knots, or 15m/s. That's an excellent clip for something that large in the water... But it's peanuts for space. A spaceship may be able to accelerate at up to 20G's for upwards of a minute. That's 9.8m/s*60 seconds, or 588 meters/s.

Designing a structure that can withstand 100,000 tons @ 15m/s is an entirely different critter than designing one that can withstand 5 million tons @ 600 m/s.

[Astral]

Looks more than a bit amazing, with a lot of raw potential. Kickstarted a bit, but unfortunately my funds are too low to do more than $10.

[Anvilfolk]

Screw it, supporting this as well

You must know a lot more tweed-wearing cavers than I do! Nothing about my description is particularly /amazing/, but a lot of the hobbies I pick up are not particularly common. Oh, and you'll have to work a lot harder to offend me!

I'm in academia, so know a lot of eccentrics, quite a few cavers, and you're talking to a (hobbyist) blacksmith. I think the people I know just don't make a fuss about it

[Girlinhat]

Ramming another ship in space is basically an insta-kill for both vehicles.  When you consider the size and speed of space, then any contact is lethal.  Look at the current space shuttles - something half the size of a dime can punch a hole straight through a ship!  Add layers of exotic armor to the ship, and make the half-dime into a MAC truck with a titanium skeleton.

The real trick is to making sure that your whole vehicle isn't destroyed when you hit the enemy.  Just let the disposable parts be destroyed.  That should be available to the player.  If they just run headlong, then let them die.  If they're smart and they design a weapon system that works, then let it work!  So long as the game doesn't say "Ship A hit Ship B, destroy them both" then the players will develop ramming weapons.

[Geneoce]

Shields

I think I'll make shields magnetic fields - This seems like the most plausible real-world phenomenon that corresponds to sci-fi shields.

If your going to add in magnetic fields, consider using plasma contained in magnetic fields! Looks cooler when you press the "shield on" button

[Girlinhat]

That would provide a good deterrent for ramming or boarding attacks, and it might mitigate energy attacks, but projectiles will just turn into plasma-charged projectiles as they hit your hull.

[dirkson]

Thanks for the support from Astral and AnvilFolk!

I'm in academia, so know a lot of eccentrics, quite a few cavers, and you're talking to a (hobbyist) blacksmith. I think the people I know just don't make a fuss about it

Yup, that'll be it. These get much rarer outside of academia. Try to keep these friends! They sound like awesome people.

Can I see some examples of your work? I've always been envious of people who take up blacksmithing.... I wonder if I could convince the landlady to let me set up a little section of the property as a blacksmithing area.

If you're going to add in magnetic fields, consider using plasma contained in magnetic fields! Looks cooler when you press the "shield on" button

That is a BRILLIANT idea. It makes them visible, so that they can look more like sci-fi shields, provides an excellent way to get rid of waste heat (Pump it into the shields as plasma!) and provides some protection against lasers (Lasers will heat up the plasma as they pass through, losing some energy.

Projectiles may be deflected slightly by the field too, depending on what they are, how fast they're moving, and how strong the shield is.

This may backfire while ramming, though - While you'll start out shoving a field of plasma into your enemy, physics tell us the force from the enemy ship impacting the shield will impart a force on the shield generator, possibly sending it shooting through your ship and out the back end. This will cause it to lose contact with power lines, causing a shield failure and spraying the entire area in plasma.

Cheers,
-Dirk

[Girlinhat]

Plasma wouldn't sustain itself very long in the void of space.  As soon as the shields drop, then you'd get a burst of heat in the immediate area and then it evaporates.  If ships use thermal scanners, it might be a way to fry their sensors and hide, maybe?

[The Fool]

You know... You can build a spike tossing ship if you assume they're going to use mag shields. Just have a heavy non-metal central spike surrounded by a breakaway material (probably butter) while the rest of the ship is metal. When you hit the shield the heat will cook the butter, toss the spike, and bounce the remaining non-breakaway part away from the ship. Mind you this is the theoretical ideal, but I wouldn't mind trying it once flight and combat are introduced.

[Sirus]

This looks rather cool. Watching, and maybe I'll have some suggestions later (once I pick my jaw off the floor )

[dirkson]

Plasma wouldn't sustain itself very long in the void of space.  As soon as the shields drop, then you'd get a burst of heat in the immediate area and then it evaporates.  If ships use thermal scanners, it might be a way to fry their sensors and hide, maybe?

You know... You can build a spike tossing ship if you assume they're going to use mag shields. Just have a heavy non-metal central spike surrounded by a breakaway material (probably butter) while the rest of the ship is metal. When you hit the shield the heat will cook the butter, toss the spike, and bounce the remaining non-breakaway part away from the ship. Mind you this is the theoretical ideal, but I wouldn't mind trying it once flight and combat are introduced.

would it be possible to make sort of ramming-frag-torpedoes?

you launch it from the ship, it splits into a mass of fragments on the way to the ship, and then spreads damage everywhere, as well as serving as a good anti-fighter weapon...

I can now imagine people, over time, changing tactics. big ships at first, then they find that small ships are harder to target, so more fighters, then they make flak weapons, which can easily hit and destroy small ships, but cause little damage to armour, so armoured ships become popular, then they use AP weapons to take these out, they will struggle to hit small ships, so...

I love all these weapon suggestions! And most wouldn't even require I do anything different from what I currently plan!

I imagine that, yes, we're likely to see a progression on the average type of ship designed, and that it's likely to shift a little from solar system to solar system - Competing cultures, basically, with a decent amount of mixing. Should be really interesting to watch!

I've actually already seen it in the current stages - One guy builds a big ship with lots of weapons. This offends another guy, who has to build a bigger, better-designed ship with MORE weapons. Insolent and CommanderKeen went back and forth 3-4 times before settling into an uneasy truce underneath's Insolent's gargantuan S.S. Queen Mary Ann.

This looks rather cool. Watching, and maybe I'll have some suggestions later (once I pick my jaw off the floor )

Thanks!

Cheers,
-Dirk

[Shades]

Which is a huge thing, since there's absolutely infinite amounts of data to send (like the temperatures and whatnot).

Most of that is deterministic though so wouldn't need to be sent after the initial event.

[10ebbor10]

You could also use different types of plasma weapons. Since plasma is ionized, it can conduct power, and reacts to magnetic fields. Supposedly you'd get in range you could fling plasma at the enemy ship to shortcircuit their systems. Another way to do it is create a missile with a ridicously powerfull shield to keep the plasma together, and then drop that on the enemy ship.

Problem with using plasma shielding is that it's serves as a semi insulator (Keeping the heat trapped near your ship) and that it is rather noticeable. Fun fact is that it obscures your ship somewhat, so you could use it to make your enemies think your ship is  much larger than it actually is by putting an overpowered shield on top of it.

[Anvilfolk]

Which is a huge thing, since there's absolutely infinite amounts of data to send (like the temperatures and whatnot).

Most of that is deterministic though so wouldn't need to be sent after the initial event.

Good point. Do floats or doubles behave the same in all hardware? Also, timing is a big issue. It's impossible to absolutely synchronise clocks, and if a ship gets hit at a precise point in time, it might cause different heats in different computers due to lag. Maybe at some point all computers just need to synchronise heat?

[Shades]

Good point. Do floats or doubles behave the same in all hardware? Also, timing is a big issue. It's impossible to absolutely synchronise clocks, and if a ship gets hit at a precise point in time, it might cause different heats in different computers due to lag. Maybe at some point all computers just need to synchronise heat?

Floats or doubles no, but if your relying on deterministic networking for multiplayer stuff you stick rigidly with integers because you can trust the maths not to start having errors creep in. As to synchronisation you make sure all events are associated with a tick* and so each machine will know when to apply a given event and will be in the same state during that time. If a given machine can't keep up it can either be dropped or warn peers/server and the whole game slows down to the slowest machine speed.


* You don't have to tick that often internally, 10 times a second is plenty more than people notice. Obviously this isn't the same as framerate but as you should have graphics and engine decoupled as much as possible this isn't a problem. Any Frames between ticks can just interpolate world data, if they are slightly out between machines it's not a big issue as within 100ms you'll have another identical tick.